About this Blog
About The Author
Email Me

RealClearPolitics HorseRaceBlog

By Jay Cost

« Race, Realignment, and the Election of 1948 | HorseRaceBlog Home Page

Does Jon Stewart Influence Public Opinion?

As a politics nerd, I have several subscriptions to very nerdy academic journals. One of them is Political Behavior. I like to keep the nerdiness quotient of this blog reasonably low, so it is rare that I actually blog on one of the articles in this journal, or any other for that matter.

But an article in the recent edition of Political Behavior caught my eye. It hooks into an ongoing discussion people have been having about the slant of political coverage, particularly over The Daily Show with Jon Stewart. Written by Jonathan Morris of East Carolina State University, the article, entitled "The Daily Show With Jon Stewart and Audience Attitude Change During the 2004 Conventions," is worth a discussion.

In recent years, scholars have found that late-night entertainment programs can have political effects - educating viewers and increasing interest in politics among the non-active. In particular, there has been a good bit of research into The Daily Show. Morris says, "Overall, the consensus of this research is that The Daily Show does have the potential to influence political discourse as well as overall attitudes."

Stewart, of course, has made no secret of his leanings toward the left - and others have noted that his show tends to skewer Republicans and conservatives more harshly than Democrats and liberals. Anybody who watches the show will probably see this tendency, but what of it? I never thought it mattered much, personally. Morris does a fair job of summarizing my view:

What if The Daily Show and Jon Stewart are friendlier toward Democrats and more critical of Republicans? Why should such a tendency be considered significant? The program is a half-hour, entertainment-based, talk show that openly mocks its own credibility as "fake news."

This has been my thinking for a while. I didn't think Stewart was particularly fair to Jim Cramer - for instance - but so what? Many who thought that interview was consequential are media types who don't have a good grasp of public opinion. And anyway, they're looking for copy - so invariably they'll make more of something like that than it deserves. In the ocean of American politics, that was just a tiny ripple. Does it really matter all that much?

Morris's analysis has given me some pause. He looks at The Daily Show's coverage of the two political conventions in 2004 to answer two empirical questions:

(a) whether or not The Daily Show's brand of humor during the 2004 party conventions was more sharply pointed at Republicans than at Democrats, and
(b) whether or not The Daily Show's audience during this time became more critical of the Republican nominee for President.

The conventions are a good way to test this hypothesis because there are very few differences between the events. They are scripted party infomercials that have a nearly-identical format.

Morris finds that the number of Daily Show jokes per convention was similar, but their content was different. Specifically:

Humor of a physical nature, which tends to highlight humorous appearance, momentarily odd behavior, incidental errors, or political miscalculation was employed at the Democratic Convention with much greater frequency than the Republican Convention. That is, political awkwardness was a much more common theme during the Democratic Convention. During the Republican Convention, on the other hand, political awkwardness was much less themed in favor of humor that focused on policy and character. Both policy and character shortcomings were exploited for laughs with significantly more frequency during the Republican Convention. [Emphasis Mine]

So, in other words, Democrats were covered in a "more light-hearted fashion," identified as pandering or bumbling. But the humor at the RNC took a "sharper edge," focusing on policy failures and character flaws. Thus, Morris answers his first question in the affirmative - The Daily Show was "more sharply pointed" at the GOP.

To answer his second question, he relies on the National Annenberg Election Study, which interviewed the same respondents before and after the conventions 2004. He examines in particular those who identified The Daily Show as one of their regular viewing choices to see if watching Stewart changed their views. Controlling for basic demographic and political factors (like race and partisanship), Morris finds that favorability toward President Bush after the RNC was negatively correlated with watching The Daily Show. In other words, all else being equal, watching Jon Stewart during the RNC made one less likely to be favorable of Bush. Morris also tested for other late night comics, as well as talk radio, newspaper, cable and network news - and did not find such a correlation anywhere else (though watching Letterman during the DNC was correlated with less favorability toward George W. Bush).

In particular, Daily Show viewers were less likely to say that "Bush is easy to like as a person," "Bush is trustworthy," and "Bush says one thing and does another" after the RNC. [Morris finds a similar result for Letterman: watching Letterman during the RNC made one less trusting of Bush.]

Now, before we go off proclaiming that this is definitive evidence of media bias influencing public opinion, some caveats are in order. First, Morris tested many types of media - the late night comedians, talk radio, cable and network news, and the newspapers. Generally speaking, few of these media had any appreciable effect on views about the candidates. Instead, the two biggest factors were partisanship and the views one had going into the conventions. In other words, the net effect of the conventions was mostly to have one's preexisting views reinforced. Beyond Jon Stewart [and to a lesser extent David Letterman and talk radio], the media had very little influence on its viewers. Broadly speaking, this cuts against the idea that bias in news or entertainment influences public opinion.

Second, we have to be careful not to overestimate the magnitude of Stewart's effect, even if we admit he has one. It's one thing to talk about a news slant having an influence on people. But when we're talking about an influence on people's political choices, we have to keep in mind the appropriate scale. Last week The Daily Show pulled in 1.6 million viewers. That's fewer than Adult Swim on Cartoon Network and just 1.2% of the total number of people who voted for President last November. Additionally, more than half of the core audience for The Daily Show identified itself as Democrat during both conventions in 2004, and only between 20% and 40% labeled themselves Independent, the political group whose votes are most up for grabs. So, we're talking about an even smaller fraction of the total population that might actually be swayed. And, from the looks of it, those who are otherwise inclined to the GOP click away from Comedy Central at 11 PM - meaning that while Stewart might be skewering Republicans most harshly, he is basically preaching to the converted.

So, I think the answer to the title question has to be an extremely qualified yes. Those who consistently view the Daily Show probably can be swayed to the left - but this is a relatively small slice of the public, which means that the effect on political outcomes is probably quite small. On top of that, the audience seems to be pretty self-sorted, anyway. So, the effect in 2004 might have been real on some people, but in the grand scheme of things it was quite modest.

Update: A few weeks ago, Gary Andres also wrote on the Morris article about Jon Stewart for The Weekly Standard online. His analysis is quite good. I encourage you to check it out!